Results a. Descriptive statistics
9
trading market dictate the absence of efficient form of capital market. Return on asset is defined as the ratio of earning before interest, extraordinary item, and taxes to total asset as of 2002. The
study use family as the unit of analysis and therefore we aggregate the individual shareholding of family members of controlling owners to construct controlling family ownership. Capital Market
Law 1995 article 1 states that family affiliation refers to the relationship by marriage and blood both to second degree vertically and horizontally. Controlling shareholders ownership is defined
by simply accumulating the cash-flow right of their immediate ownership proportional to total number of common shares. A 20 cut-off is used in differentiating between family-controlled
and non-family-controlled firm
1
. Foreign shareholders and domestic blockholder are defined as institutional shareholders that are independent of controlling family, with at least 5
shareholding of the firm. Foreign and domestic blockholder ownership is defined as the proportion of their shareholding to total number of outstanding shares. The fraction of
independent directors is defined as the ratio of independent directors to total numbers of directors.
4. Results 4. a. Descriptive statistics
Table 1 presents descriptive statistics and the correlation of variables of interest. Overall, the correlation coefficients between independent variables are relatively low indicating that there is
no presence of multicollinearity problem, except for controlling family ownership FMLY and foreign shareholding FRGN. It is not surprising given that family ownership, foreign
1
See for example LaPorta, Lopes-de-Silanes, and Sheifler 1998 and Claessens, Djankov and Lang 2000. However, it should be noted that this cut-
off point is best viewed as “researcher discretionary” as, to date, there is no theoretical work justifying this point.
10
shareholding, and domestic blockholding are mutually exclusive in nature. The average shareholding by controlling family is 57 ranging from 0 minimum to 98 maximum
2
. Further analysis reveals that controlling owners hold more than 50 of corporate shares in 103
firms 70 of the sample. Controlling owners are absent in only 21 firms, representing 12 of total sample. This description confirms the work of Lukviarman 2004 documenting the
ownership concentration by controlling family in 70 Indonesian listed firms during 1994 to 2000. The presence of controlling shareholders is evidence in 88 of the sample 167 firms.
Diffused ownership, where minority investor cumulatively own more than 80 of corporate shares, is found in only 5 firms representing 3 of total sample that supports that of Claessens,
Djankov, Lang 2000 revealing the prevalence of concentrated ownership firms and that only small numbers of Indonesian firms have dispersed ownership structure.
Insert table 1 about here -
The average foreign investor‟s shareholding is 11 with 0 as the minimum and 96 as the maximum. However, only one third of firms exhibit the presence of foreign shareholders where
such shareholders are absent in most listed firms 135 firms representing 71 of the sample. Foreign investors own 50 or higher of the corporate shareholding in 20 firms representing 11
of the sample. Further analysis reveals that 5 of the listed firms are jointly owned by foreign and domestic investor with 50 cumulative shareholding or higher. ADB 2001 contends that
such a coalition provides foreign investor with access to local market and political connection. Domestic blockholders are found in only 11 of the sample 22 firms with an average
2
In all sample, the immediate owner of a firm is another company of particular business groups owned by the same controlling owners. This ownership structure, so-called pyramidal ownership, is consistent with the finding of
Claessens et al 2000, 2001. In some occasion, the firm is jointly owned by several families who form the partnership to control the firms. However, this joint ownership is a floating coalition, instead of permanent coalition,
where the partnership changes in other firms.
11
shareholding of 2.5 ranges from 0 minimum to 49 maximum. Of these firms, the highest frequency falls into 5 to 20 ownership category. Eighty-eight percent of the sample
168 firms displays the absence of domestic blockholder suggesting that the absence of external large shareholders, who are independent of controlling shareholders, is a norm in Indonesian
listed firms.
Controlling family ownership exhibits a negative correlation with firm performance ROA2, suggesting that higher family ownership is associated with lower firm performance. Foreign
ownership is positively correlated with firm performances, indicating that foreign-controlled firm is more likely to have superior performance as compared to family-controlled firm. The
correlation coefficient between the proportion of independent directors and firm performance is insignificant. Taken together, the findings indicate that different owners have different
association with firm performance. In other words, the identity of large shareholders does matter in predicting organizational outcome.
The proportion of independent directors IDPD is positively correlated with domestic blockholder DOM, suggesting that higher shareholding by domestic independent large
shareholders is associated with a higher fraction of independent directors. By contrast, family ownership and foreign shareholders are insignificantly associated with the fraction of
independent directors. Again, the findings reveal that different owners pursue different choices of governance mechanisms. The proportion of independent directors is positively correlated with the
proportion of independent directors serving on audit committee AUDC, indicating that an outsider dominated board tends to have a higher proportion of independent directors serving in
12
audit committee. However, the proportion of independent directors and audit committee are insignificantly correlated with firm performance is insignificant indicating that monitoring by
independent board is less likely to mitigate agency problem in Indonesia.